JP5284138B2 - Image processing apparatus and image processing method - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing method. In particular, the present invention relates to a configuration in which multi-valued data of multiple colors is quantized using a table that receives these multi-valued data as input, and a quantization value determined by correlating the quantized values of multiple colors at each lattice point It relates to processing.

  As a quantization method for converting multi-value data into binary data, an error diffusion method and a dither method are widely known. In the dither method, the data is quantized into data having a smaller number of gradation values than the multi-value data, such as binary data of each pixel, based on the magnitude relationship between the threshold value given to each pixel by the dither matrix and the value of the multi-value data. Is.

Similarly, the error diffusion method quantizes based on the magnitude relationship between the threshold value and the value of the multi-value data, diffuses the error between the threshold value and the multi-value data at the time of quantization to the surrounding pixels, and then performs the quantization. The multi-value data of the pixel to be converted is corrected. For example, the density (multi-value data value) of the target pixel P for obtaining binary data as quantized data is v, and the densities of the peripheral pixels P0, P1, P2, and P3 that diffuse the error of the target pixel P are v0. , V1, v2, v3, and T is the threshold value. Further, binary data O determined by the pixel P when the weights when the error E due to the binarization of the pixel of interest P is diffused to the peripheral pixels P0, P1, P2, and P3 are W0, W1, W2, and W3. Is
If v ≧ T, O = 1, E = v−Vmax;
If v <T, O = 0, E = v−Vmin;
Here, Vmax is the maximum density and Vmin is the minimum density.
It is represented by Also, the density of the surrounding pixels as a result of error diffusion is
v0 = v0 + E × W0;
v1 = v1 + E × W1;
v2 = v2 + E × W2;
v3 = v3 + E × W3;
It is expressed. The weighting factors in these equations are, for example, W0 = 7/16, W1 = 1/16, W2 = 5 / 16W3 = 3/16.

  As for the above-described dither method and error diffusion method, in general, the dither method can perform high-speed processing, and the error diffusion method has an advantage of excellent image quality. In recent years, in order to enrich the gradation expression, the quantized data is not limited to binary values but is set to three or more values. In the dither method, by preparing a plurality of dither matrices, and in the error diffusion method, by setting a plurality of threshold values, quantized data of three or more values can be obtained.

  However, in the quantization as described above, when multi-value data of a plurality of colors such as cyan, magenta, and yellow is quantized, there is a problem caused by independently quantizing each multi-value data. That is, the images recorded based on the respective quantization results have good visual characteristics, but the images recorded on the basis of the respective quantized data do not necessarily have good visual characteristics. You may not have it. This is because each multi-value data is quantized independently, and superposition of multi-value data of a plurality of colors is not considered.

  On the other hand, Patent Document 1 describes that a lookup table is used and quantization by error diffusion of multi-value data of a plurality of colors is performed in association with each other. Specifically, the quantization values for each of a plurality of colors are set at each lattice point of the table, and the quantization values corresponding to the multi-value data of each color are obtained using this table, and these quantization values are calculated in the table. It is described that correction is performed in accordance with a density region for each perceived density defined. According to this, good visual characteristics can be obtained in an image in which a plurality of colors are overlaid.

  By the way, in a recording apparatus such as a color inkjet printer, recording is performed using four types of ink, which are basic inks of cyan ink (C), magenta ink (M), yellow ink (Y), and black ink (K). Widely known. On the other hand, in addition to the above four colors, ink with a relatively low colorant density such as dye (light ink) is used for recording of higher image quality, for example, graininess reduction in the highlight area, and shadow area. It is known to perform recording that achieves both high density and low density. There is also known an apparatus that performs recording with a plurality of dot sizes for recording with different sizes of ink droplets ejected for the same color, and similarly performs high-quality recording. That is, in these apparatuses, an image is formed with dots (light ink or small dots) having a relatively low optical density in a highlight portion (low density area of recording density), and a shadow area (high density area of recording density). Then, an image is formed with dots (dark ink or large dots) having a relatively high optical density.

  Further, in recording not only color images but also monochrome images, high image quality equivalent to silver halide photography is required, and gradation is particularly important. In this case, in a configuration in which a monochrome image is formed by mixing colors of C, M, and Y inks, if slight variations in C, M, and Y inks occur, for example, variations in the landing positions of inks and ink droplet sizes, the image In this case, there is a problem that color shift occurs and the image quality is deteriorated. On the other hand, in the configuration in which an image is formed using only black ink, there is no variation in the landing position and ink droplet size described above, and image quality deterioration such as color shift does not occur. In view of the above, a recording apparatus that determines whether an image to be recorded is a color image or a monochrome image and switches the type of ink to be used according to the determination result has been proposed. That is, C, M, and Y inks are mainly used when recording color images, and K ink is mainly used when recording monochrome images. Furthermore, in the monochrome image recording, as in the color recording described above, the use of a dot having a relatively low optical density and a dot having a relatively high optical density achieves a reduction in graininess in the highlight area and a high density in the shadow area, thereby improving the photographic image quality. Recording is possible. As one configuration for realizing this, there is known an apparatus that performs recording using black ink (black ink: K ink) having a high color material density and black ink (grey ink: GY ink) having a low color material density. ing.

  FIG. 1 is a diagram for explaining how to use K ink and GY ink. In the highlight portion (region I shown in FIG. 1) in the vicinity where the density level indicated by the input data is the lowest, recording is performed using only GY ink. Further, as the input level increases from this area, the amount of GY ink shot (density output level) increases (area II). Thereafter, when the input level exceeds a predetermined value, the K ink starts to be used, and a relatively large amount of GY ink is maintained (area III). As a result, K dots can be sparsely arranged with the GY density increased. In the region where the input level is close to the maximum (region IV), recording is mainly performed using only K ink.

  However, in the above configuration, when the difference in the optical density of dots recorded by the GY ink and the K ink is large, the following problem may occur. In the region III where the dark dots by the K ink are sparsely arranged, the density that can be increased by the light dots by the GY ink is not sufficient, and as a result, the dark dots may be visually recognized to give a grainy feeling. . As for this, if the density of light dots is increased to reduce the difference in recording density, this problem hardly occurs, but the effect of reducing the graininess of the highlight portion and the maximum density cannot be achieved sufficiently.

  On the other hand, this problem can be alleviated by recording with three or more types of dots in which dots having intermediate optical densities are added in addition to GY and K dots. For example, recording is performed using three types of dots including a large dot made of high-density black ink (K ink), a large dot made of low-density black ink (GY ink), and a small dot made of low-density black ink (GY ink). To do.

  FIG. 2 is a diagram for explaining how to use the above three types of dots, and is similar to FIG. As shown in FIG. 2, in the highlight portion (region I), recording is performed only with small dots of GY ink (hereinafter referred to as GY small dots) that are the dots with the lowest optical density. Increase the amount (area II). Thereby, it is possible to reduce graininess in a relatively low density region. When the input level exceeds a predetermined value, large dots of GY ink (hereinafter referred to as GY large dots) are sparsely arranged (area III). As the input level further increases, the amount of GY small dots is reduced, while the GY large dots are increased to increase the overall density (region IV). When the input level further increases, large dots of Bk ink (hereinafter referred to as Bk large dots) begin to be used sparsely (area V), and in areas close to the maximum input level (area VI), only large Bk dots are used. Use to record.

  In this way, by performing recording using three types of dots, the density difference between the GY small dot and the GY large dot, and the GY large dot and Bk are compared with the case where recording is performed using only two types of dots. The density difference from the large dots can be reduced. This suppresses the increase in graininess in the halftone area, particularly in the areas III and V where the dark dots start to enter, while reducing the graininess in the highlight part of the recorded image and reducing the high density of the shadow part. It can be realized well. In addition, as described above, the number of types of dots to be formed can be increased to reduce graininess, particularly in the halftone area, not only in the case of recording monochrome images but also in color recording using color inks. is there.

  However, increasing the types of ink or dots naturally increases the size and complexity of the printing apparatus and the associated cost increase.

JP 2003-224730 A JP 2000-108420 A

  In view of this point, there is a demand for a recording apparatus that can minimize deterioration of graininess particularly in a halftone area even with a small number of ink types or dot types. For example, the quantization technique described in Patent Document 1 described above can be used for this problem. In other words, this technique stores each quantized value corresponding to multi-value data of a plurality of colors at a grid point of the table, and the quantized value of each grid point is stored in a density region defined based on the perceived density in the table. Quantization is performed using a table with corresponding values. According to the quantization using this table, first, since the quantization value is adjusted corresponding to the density region based on the perceived density, for example, in the region where dark dots are sparsely arranged, The arrangement of dark dots and light dots is determined so as to satisfy the perceptual density. Therefore, in a region where dark dots are sparsely arranged, it is possible to avoid a state where the density that can be increased by the light dots is not sufficient, and as a result, it is possible to reduce the graininess due to the dark dots. Second, as described above, since the quantized values correspond to the density regions and the quantized values of the shades are associated with each other, the error of the quantized values obtained by this table is indirectly determined by the multi-value data of the other party. It will also affect. As a result, the light and dark dots are arranged separately from each other due to the characteristics of error diffusion. As a result, it is possible to reduce the probability that the light and dark dots are overlapped or arranged close to each other, and to reduce the graininess due to the light and dark dot cluster. As described above, for the sake of convenience, a quantization method that performs error diffusion processing of a plurality of multi-value data using a table in relation to each other and separates different types of arranged dots as much as possible will be described below. This is called “quantization (separation error diffusion)”. Therefore, the separation quantization may be performed not only using the method described in Patent Document 1, but also using another method, for example, the dither method described in Patent Document 2. This document describes that when quantizing data of multiple types of ink, the quantization of a certain type of ink data minimizes dot overlap related to the previously quantized type of data. ing. As described above, in applying the present invention, any kind of quantization can be used as long as dots of different types formed based on quantization data are arranged so as to be mutually dispersed. It may be a method.

  As described above, by using the technique of Patent Document 1, even when fewer types of dots are used, it is possible to suppress deterioration of graininess particularly in a halftone area.

  However, the technique described in Patent Document 1 has a problem that when the type of multi-value data to be quantized, that is, the type of data referring to a table increases, the table size increases by the power of the number of types. For this reason, when the technique of Patent Document 1 is applied to a recording system in which the types of ink and dot size to be used are, for example, three or more types, the table size becomes enormous, leading to an increase in size and cost of the apparatus.

  The present invention has been made to solve the above-described problem, and image processing that enables reduction of graininess, particularly in a halftone region, by quantization using a table without causing an increase in table size. An apparatus and an image processing method are provided.

In the present invention Therefore, an image processing apparatus for recording a first dot optical density are different, the image is formed on the second dot and the third dot over the recording medium in the same color, the image color A color for generating first multi-value data corresponding to the first dot, second multi-value data corresponding to the second dot, and third multi-value data corresponding to the third dot based on the component data When the values indicated by the decomposing means and the first, second, and third multi-value data are not 0, the third multi-value data is used by using the first multi-value data and the second multi-value data. The first quantized data for forming the first dot is generated without using the third multi-value data by using the first multi-value data and the second multi-value data. generating a second quantization data for forming a Quantizing means that, characterized in that obtaining immediately a.

The first dot different optical density respectively in the same color, an image processing method for recording an image by forming a second dot and the third dot on the recording medium, based on the color component data of the image A color separation step of generating first multi-value data corresponding to the first dots, second multi-value data corresponding to the second dots, and third multi-value data corresponding to the third dots ; If none of the values indicated by the first, second, and third multi-value data is 0, the first multi-value data and the second multi-value data are used and the third multi-value data is not used. Generating the first quantized data for forming one dot and forming the second dot using the first multi-value data and the second multi-value data without using the third multi-value data; quantization step and for generating second quantized data Characterized in that it have a.

  According to the above configuration, the number of multi-value data referring to a table used for quantization can be at most 2, and an increase in table size can be suppressed.

It is a figure explaining how to use K ink and GY ink when recording an achromatic color. It is a figure explaining how to use three types of dots, a large dot using K ink, a large dot using low density black ink, and a small dot using low density black ink when recording an achromatic color. 1 is a block diagram illustrating a recording system including an inkjet recording apparatus and a host computer according to an embodiment of the present invention. It is a perspective view which shows the outline of the mechanical structure of the inkjet recording device shown in FIG. FIG. 5 is an exploded perspective view for explaining details of the head cartridge shown in FIG. 4. 2 is a cross-sectional view schematically showing a schematic configuration of a main part of a recording head constituting the cartridge. FIG. It is a block diagram which shows the structure for the image processing which concerns on one Embodiment of this invention along the series of processes. It is a figure which shows the relationship between the monochrome gradation value G obtained by the gray scale conversion part shown in FIG. 7, and the output data from a gradation correction process part. It is a flowchart which shows the quantization process which concerns on 1st embodiment of this invention. It is a figure explaining LUT used for the isolation | separation error diffusion process in the said quantization. It is a figure explaining the density | concentration area | region in the said LUT. It is a figure which shows the dot pattern produced | generated based on the quantization data obtained by the said quantization process. It is a flowchart which shows the monochrome mode quantization process which concerns on 2nd embodiment of this invention. It is a flowchart which shows the monochrome mode quantization process which concerns on the modification of said 2nd embodiment. It is a figure which shows the relationship between the input level of a cyan component and the output value from a gradation correction process part based on 3rd embodiment of this invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(Embodiment 1)
FIG. 3 is a block diagram showing a recording system including an inkjet recording apparatus 301 and a host computer (hereinafter also referred to as host apparatus) 302 according to an embodiment of the present invention. The recording device 301 and the host device 302 are connected by a known communication means and can communicate with each other. The user can output a desired image from the recording device 301 by accessing the host device 302. At this time, the host apparatus 302 executes various processes as will be described later on the image data to be recorded, and transmits the recording data obtained as a result to the recording apparatus 301.

  FIG. 4 is a perspective view schematically showing the mechanical configuration of the ink jet recording apparatus 301 shown in FIG. 3, and shows a state where the front cover of the apparatus is removed and the inside of the apparatus is exposed. In the figure, 1000 is a replaceable head cartridge, 2 is a carriage unit for detachably holding the head cartridge 1000, and 3 is a holder for fixing the head cartridge 1000 to the carriage unit 2. The head cartridge 1000 is integrally provided with a recording head provided with an ejection port for ejecting ink. When the head cartridge 1000 is mounted on the carriage unit 2 and then the cartridge fixing lever 4 is operated, the holder 3 is interlocked and the head cartridge 1000 can be pressed against the carriage unit 2 and fixed. In addition, the head cartridge 1000 is positioned by this pressure contact, and at the same time, a contact between a required signal transmission electrical contact provided on the carriage unit 2 and an electrical contact on the head cartridge 1000 side is performed.

  Reference numeral 5 denotes a flexible cable for transmitting an electrical signal between the apparatus control unit and the carriage unit 2. Reference numeral 6 denotes a carriage motor serving as a drive source for reciprocating the carriage unit 2 in the main scanning direction, and reference numeral 7 denotes a carriage belt that transmits the drive force to the carriage unit 2. Reference numeral 8 denotes a guide shaft that extends in the main scanning direction to support the carriage unit 2 and guide its movement. Reference numeral 9 denotes a transmission type photocoupler attached to the carriage unit 2, and 10 denotes a light shielding plate provided near the carriage home position. Thereby, when the carriage unit 2 reaches the home position, the light shielding plate 10 blocks the optical axis of the photocoupler 9 so that it can be detected that the carriage unit 2 is located at the home position. 12 is a home position that includes a recovery mechanism such as a cap member that caps the recording head in the head cartridge 1000, a suction mechanism that performs suction inside the discharge port through the cap member, and a member that wipes the recording head discharge port surface. Indicates a unit.

  Reference numeral 13 denotes a discharge roller for discharging the recording medium. The recording roller is sandwiched in cooperation with a spur roller (not shown), and is discharged out of the recording apparatus 301. Further, the recording medium is conveyed by a predetermined amount in the sub-scanning direction by a line feed unit (not shown). A paper end sensor (PE sensor) (not shown) is provided on the conveyance path of the recording medium, and the leading end and the trailing end of the recording medium can be detected.

  FIG. 5 is an exploded perspective view for explaining the details of the head cartridge 1000. In the figure, reference numeral 15 is a replaceable ink tank for storing Bk (black) ink and GY (gray) ink, and 16 is an exchange for storing ink of each colorant of C (cyan), M (magenta), and Y (yellow). Each possible ink tank is shown. Reference numeral 17 denotes an ink supply port of the ink tank 16 which is connected to the head cartridge 1000 and supplies ink, and 18 denotes an ink supply port of the ink tank 15 similarly. The ink supply ports 17 and 18 are connected to the supply pipe 20 and configured to supply ink to the recording head 21. Reference numeral 19 denotes an electrical contact connected to the flexible cable 5 and configured to transmit a signal based on the recording data to the recording head 21.

  In FIG. 5, eight lines shown on the front surface of the recording head 21 indicate each nozzle row including a plurality of nozzles that eject inks having different ink colors and ink droplet sizes. Specifically, black ink droplet (large; Bk), cyan large ink droplet (C), cyan small ink droplet (c), magenta large ink droplet (M), magenta small ink droplet (m), yellow ink droplet (Y) , Large gray ink droplets (GY) and small gray ink droplets (gy) are ejected. In this embodiment, the large ink droplet is about 5 pl, and the small ink droplet is about 2 pl.

  FIG. 6 is a cross-sectional view schematically showing a schematic configuration of a main part of the recording head 21. In the figure, reference numeral 4000 denotes a base plate, which includes common liquid chambers 5101, 5103, 5105, 5107, 5109, 5111, 5113 and 5115 for receiving five types of ink for ejection. Each common liquid chamber is connected to supply paths 5102, 5104, 5106, 5108, 5110, 5112, 5114, and 5116 formed by anisotropically etching the back surface of the heater board formed in the semiconductor manufacturing process. . Thus, the ink path groups corresponding to the ink discharge heater groups can be separated or partitioned so as not to cause mixing of different color inks.

  Reference numeral 4001 denotes a heater board for black ink and large cyan ink droplets, and reference numeral 4002 denotes a heater board for small cyan ink droplets and large magenta ink droplets. Reference numeral 4003 denotes a heater board for magenta small ink droplets and yellow ink, and reference numeral 4004 denotes a heater board for large gray ink droplets and small gray ink droplets. Each heater board is provided with heaters 5004 and 5006 for discharging. Reference numerals 5001, 5002, 5003 and 5004 denote orifice plates formed with ink flow paths and nozzles, and are usually formed of a heat-resistant resin.

  For example, taking black ink as an example, the black ink supplied from the liquid chamber 5102 is supplied to the ink flow path in the vicinity of the ejection heater groups 5004 and 5006. When voltage pulses are applied to the individual heaters 5004 and 5006 in accordance with the recording signal, ink is ejected as droplets from the ejection ports 5005 and 5007 in the direction of the arrow due to the foaming energy of film boiling caused by rapid heating of the heaters. Is done. The ejected ink droplets land on the recording medium P and form dots corresponding to the ink color and the size of the ink droplets. In the present embodiment, one ink chamber 5102 corresponds to two nozzle rows 5005 and 5007, and the left nozzle row 5005 is referred to as an even nozzle and the right nozzle, ie 5007, as an odd nozzle in the drawing.

  FIG. 7 is a block diagram showing a configuration for image processing according to the present embodiment along a series of steps. Specifically, these image processes are executed by an image processing configuration having a CPU, a ROM, a RAM, and the like of each of the host apparatus 302 and the recording apparatus 301 shown in FIG.

  Image data created by the application 701 of the host device 302 is represented by an 8-bit luminance signal having 256 gradations of red (R), green (G), and blue (B). This image data is finally converted into quantized data used by the printing apparatus by the printer driver 3020 and is sent to the printing apparatus 301. In the printer driver 3020, first, resolution conversion (conversion to 600 ppi in this embodiment) is performed by a resolution conversion unit (not shown). Next, the determination unit 702 determines whether the recording is a monochrome image output or a color image recording output. In the present embodiment, the user selects monochrome image output or color image output by an operation via the host 302. Of course, other known methods such as determining whether image data created by an application is a monochrome image or a color image based on the image data can be applied.

When the monochrome image output is selected, first, the gray scale conversion unit 703 performs gray scale conversion of the image data created by the application. This gray scale conversion is a process of changing R, G, and B data, each of which has 8 bits and 256 gradations, to 8 bits data of monochrome 256 gradations having only lightness information. Various conversions are known, but in this embodiment,
Monochrome gradation value G = (WR × R + WG × G + WB × B) (rounded off to the nearest decimal point)
Here, weighting components: WR = 0.299, WG = 0.487, WB = 0.114.
Next, the color separation processing unit 704 converts the monochrome gradation 8-bit data into 8-bit data corresponding to the type of ink and dot size used in the printing apparatus. That is, depending on the combination of ink and dot size, black dots (Bk dots), gray large dots (GY dots), gray small dots (gy dots), cyan small dots (c dots), magenta small dots (m dots) , Yellow dots (Y dots) can be formed, and these 8-bit data are generated. These color separation data correspond to the combination of the ink type and the ink droplet size described above with reference to FIG. It is. The color separation processing unit 704 performs the processing by referring to a color separation table prepared in advance. This color separation table includes 17 monochrome gradation values G of 0, 16, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, and 255, respectively. Correspondingly, values of Bk, GY, gy, Y, c, m are stored. In a specific process, when a monochrome gradation value G is given, the color separation table is referred to, and the values of Bk, GY, gy, Y, c, and m of the coordinate point corresponding to the gradation value are set as adjacent lattice points. Is interpolated from the values of Bk, GY, gy, Y, c, and m to obtain the value.

  In the present embodiment, the reason for outputting chromatic colors Y, c, and m while outputting a monochrome image is as follows. Generally, a recording medium such as a recording paper is white, but has a slight color. For this reason, even if achromatic Bk, Gy, and gy dots are formed, the recorded image may be slightly colored due to the original color of the recording medium. For this reason, in order to make the recorded image have the originally intended monochrome color, a small amount of chromatic ink is used to cancel the color of the recording medium. However, if too much chromatic color ink is used, color shift may occur. Therefore, the ink amount is limited to about several percent of the achromatic color ink.

  In the subsequent gradation correction processing 705, the value output from the color separation processing unit 704 is converted by the gradation correction table. That is, conversion is performed by referring to a gradation correction table in which conversion values for 256-gradation data are determined in advance for each color of Bk, GY, gy, c, and mY. In this embodiment, this input value is converted to 256 gradations of input values (= output values from color separation processing) 0 to 255, and the output range is set to 0 to 4080. As a result, finer gradation correction is possible.

  FIG. 8 shows a monochrome gradation value G obtained by the gray scale conversion unit 703 and data obtained by performing the above-described color separation processing and gradation correction processing on the monochrome gradation value G, that is, output data from the gradation correction processing unit 705. It is a figure which shows the relationship with these, and is a figure similar to FIG. Generally, it is the same as that described with reference to FIG.

  As shown in FIG. 8, GY dots start to be arranged from an input value 63 (hereinafter, value A). Also, Bk dots start to be arranged from the input value 150 (hereinafter referred to as value B). Then, although slightly, m and Y dots are arranged for color adjustment. In this embodiment, since the recording medium is bluish, the c component is not used and only the m and Y components are used.

  Referring again to FIG. 7, with respect to the value output from the gradation correction processing unit 706 as described above, the monochrome mode quantization processing unit 706 has four values (level 0, level 1, level 2, Quantization processing for conversion into 2-bit data of level 3) is performed.

  FIG. 9 is a flowchart showing the quantization processing according to the first embodiment of the present invention.

  When this processing is started in step 900, first, in step 901, for each pixel, it is determined whether or not the monochrome gradation value G obtained by the gray scale conversion unit 703 (FIG. 7) is smaller than the value B (FIG. 8). judge. If it is smaller, the process proceeds to step 902, and the same look-up table (LUT) described in Patent Document 1 is used only for the combination of data GY and gy among the data output from the gradation correction processing unit 705 (FIG. 7). The separation error diffusion for obtaining the quantized value is performed with reference to FIG.

  FIG. 10 is a diagram for explaining an LUT used for the separation error diffusion process. Specifically, FIG. 10 shows the contents of an LUT with GY and gy as inputs. The horizontal axis represents gy, the vertical axis represents the input value of GY, and the numerical values on the grid points are gy and GY, respectively. It represents the quantized value.

  As shown in FIG. 10, since the LUT performs four-value quantization for both GY and gy, the quantization representative value and the quantization threshold are set as follows. The quantized representative values are 0 for level 0, 1360 for level 1, 2720 for level 2, and 4080 for level 3. The threshold values for quantization are 688 for levels 0 to 1, 2048 for levels 1 and 2, and 3408 for levels 2 to 3. As shown in detail in FIG. 11, the threshold points described above are taken on the x-axis (gy) and the y-axis (GY). Then, a straight line connecting points having the same threshold is connected to a point that becomes the threshold of GY on the axis of x (gy) = 4080 and a point that becomes the threshold of gy on the axis of y (GY) = 4080. It can be divided into seven density regions by a straight line. And in each density | concentration area | region prescribed | regulated in this way, it adjusts so that the sum total of the quantization value of gy and GY may become equal.

  By performing quantization using the above table, first, even when there are as few as two pieces of data to be referred to (gy, GY), first, dark dots (GY) in a region where dark dots are sparsely arranged. And the arrangement of light dots (gy) are determined so as to satisfy the perceptual density. Therefore, in a region where dark dots are sparsely arranged, it is possible to avoid a state where the density that can be increased by the light dots is not sufficient, and as a result, it is possible to reduce the graininess due to the dark dots. Second, as described above, the quantized values are associated with the quantized values corresponding to the density regions, so that the gray dots are arranged separately from each other in terms of error diffusion characteristics. As a result, it is possible to reduce the probability that the light and dark dots are overlapped or arranged close to each other, and to reduce the graininess due to the light and dark dot cluster.

  When the quantization of the combination of gy and GY is completed as described above, in step 903, Bk (and m and Y respectively) are independently quantized by the error diffusion method.

  If it is determined in step 901 that the monochrome gradation value G is greater than or equal to the value B, the process proceeds to step 904. Here, using the LUT of the combination of Bk and Gy, the same quantization as the above-described quantization of the combination of gy and GY is performed to obtain quaternary data. In step 905, as in step 903, gy (and m and Y, respectively) is independently quantized by the error diffusion method.

  Regarding the quantization in the above steps 902 and 904, in the region smaller than the value B, Bk data is not used as shown in FIG. Accordingly, it is assumed that the influence of Bk data on the pixels in the region is small, and separation quantization is performed using an LUT that inputs a combination of GY and gy in order to prioritize the granularity optimization by entering GY data. . On the other hand, since Bk data is used in an area greater than or equal to value B, quantization is performed using an LUT that receives a combination of only Bk and GY. In this case, GY is combined with GY instead of gy because GY has a higher optical density than gy, so that when Bk and GY overlap, the optical density is higher when Bk and GY overlap. It is easy to recognize. That is, optimizing the combination of GY and Bk is more effective in reducing graininess.

  As described above, the number of types of ink data referring to the table can be set to 2, and an increase in the table size can be suppressed.

  Even when the output of the gradation correction processing unit 705 is 0, a dot may be formed in a pixel due to accumulation of errors from other pixels. For this reason, independent quantization processing in steps 903 and 905 is necessary.

  In addition, this embodiment relates to an example using three types of dots of gy, GY, and Bk as shown in FIG. 8, and in the description related to FIG. Even in this case, the quantization of the present embodiment described above can further reduce the graininess. That is, according to the quantization of the present embodiment, the occurrence of graininess can be effectively reduced even when the number of types of ink used is small. Therefore, the number of types of ink data referring to the table is at most two as described above. be able to. As a result, an effect that an increase in table size can be suppressed can be obtained.

  Referring to FIG. 7 again, when the determination unit 702 determines that the color output is performed, the color separation processing unit 707, the gradation correction processing unit 708, and the color quantization processing unit 709 are each processed. Since it is the same as a well-known thing, those description is abbreviate | omitted.

  FIG. 12 is a diagram showing dot patterns arranged by the dot arrangement patterning processing unit 710 (FIG. 7) based on the quantized data obtained by the monochrome mode quantization processing unit 706 (FIG. 7). Data of levels 0 to 3 of each color obtained by quantization is data for arranging 0 to 4 dots in one pixel having a resolution of 600 dpi. That is, the above four levels of density are expressed for each color by dot recording / non-recording for each area of 2 areas × 2 areas constituting one pixel area. Here, an area indicated by a circle indicates an area where dots are recorded, and an area where no circle is indicated indicates an area where dots are not recorded. As can be seen from the figure, when the input level is 0, dots are not arranged in all areas. However, the number of levels increases to 1 dot at level 1, 2 dots at level 2, and 4 dots at level 3. As the number of dots to be arranged (recorded) increases gradually.

  The print data finally binarized (dot arrangement) as described above is transferred to the print head drive circuit 711 shown in FIG. 7, and ink is ejected from the print head based on this print data. . In recording, in the 2 × 2 area of the dot arrangement pattern, the even nozzles are in charge of the upper area, and the odd nozzles are in charge of the lower area. Through the series of processes described above, 600 gradation image data of 256 gradations is recorded at a resolution of 1200 gradations of two gradations.

  In the above-described example, an example in which monochrome recording is performed using three types of dots having different optical densities has been described. However, the present invention is not limited to this. For example, with respect to black (Bk) ink dots, two types of dots, large dots and small dots, can be formed, and the present invention can be applied to a system using a total of four types of dots. That is, the monochrome gradation value G at which dots with higher optical density enter is set in advance as a predetermined value. Then, the monochrome gradation value G is compared with a predetermined value, and based on the comparison result, separation quantization is performed on a combination of a dot having an input to the quantization process and a dot having the next highest optical density. Good.

  Specifically, the gradation value at which GY large dots enter is set as value A, the gradation value at which Bk small dots enter is set as value B, and the gradation value at which Bk large dots enter is set as value C in advance. When the gradation value G <value B, GY and gy are separated and quantized. When value B ≦ monochrome gradation value G <value C, Bk small dots and GY are separately quantized. Further, when value C ≦ monochrome gradation value G, separation quantization of Bk large dots and Bk small dots is performed. The same applies to five or more types of dots.

  When this embodiment is generalized, quantized data used for recording by forming three or more types of dots having the same hue such as monochrome and different optical densities is generated. In this case, based on data indicating one color component such as monochrome gradation data, a plurality of multi-value data respectively corresponding to three or more types of dots is generated. Then, when quantizing the plurality of multi-value data to obtain quantized data having a smaller number of gradation values, the quantized data obtained by associating the plurality of multi-value data with each other by the plurality of multi-value data Quantized data is obtained by referring to the table storing. At this time, according to the value of the color component data, a table to be referred to by multi-value data corresponding to two types of dots is selected from among the multi-value data corresponding to three or more types of dots, and this table is selected from two types of dots. Quantized data is obtained by referring to multi-value data corresponding to the above.

(Embodiment 2)
FIG. 13 is a flowchart showing a monochrome mode quantization process according to the second embodiment of the present invention. The monochrome mode quantization processing of this embodiment is different from that of the first embodiment in that two data that refer to the LUT from the gradation correction processing unit 705 (FIG. 7) according to whether the Bk output is 0 or not. It is a point that makes the combination different.

  In step 1102 of FIG. 13, it is determined whether or not the Bk output value from the gradation correction processing unit is zero. When it is determined that the output value of Bk is 0, the processes after step 1103 are executed. That is, it is determined that there is almost no contribution of Bk with respect to this processed pixel, and separation error diffusion is performed in step 1103 using an LUT that receives a combination of GY and gy (1103). Thereafter, in step 1104, Bk (and m and Y, respectively) is independently quantized by the error diffusion method.

  If it is determined in step 1102 that the output value of Bk is other than 0, it is determined in step 1105 that there is a contribution of Bk to the pixel involved in the processing, and an LUT that inputs a combination of Bk and GY is used. Perform separation error diffusion. Thereafter, in step 1106, quantization by the error diffusion method is performed by gy (and m and Y respectively) alone.

  As described above, in the present embodiment, Bk, which is the dot having the highest optical density, most affects the graininess. Therefore, if there is a Bk output, it is assumed that there is a high probability that Bk is shot into that pixel. Quantization considering color overlap is performed to reduce graininess. On the other hand, when the Bk output is 0, since the probability that a Bk dot is formed in the pixel is low, quantization is performed by combining GY, which is a dot having the next highest optical density after Bk, with gy. Thereby, the granularity by GY can be reduced favorably. Even when the output of Bk is 0, it is possible that an error from another pixel accumulates to form a dot in that pixel. For this reason, even if the Bk output is 0, the Bk quantization process in step 1104 is executed.

  As described above, in the monochrome image output mode, it is determined whether or not the data value to be quantized is 0 in a system that performs recording using three types of dots having different optical densities. Then, by changing the combination of dots to be quantized based on the determination result and making each combination at most two dots, the graininess in the halftone region can be increased without enlarging the LUT size used for quantization. It becomes possible to reduce.

  FIG. 14 is a flowchart showing a monochrome mode quantization process according to a modification of the second embodiment. In the above example, monochrome recording is performed using three types of dots having different optical densities. However, the present invention is not limited to this. The present invention can also be applied to an example in which recording is performed using four or more types of dots having different optical densities.

  That is, it is determined whether the data value to be quantized is 0 in descending order of optical density among a plurality of types of dots (inks). If the data value is not 0, in order to reduce the granularity of the dot formed thereby, an LUT is used which is a combination of the dot data and the next second optical density dot data. Perform quantization. If the data value to be quantized is 0, it is determined whether or not the data value of the next highest optical density dot is 0, and the same processing is repeated.

  The example shown in FIG. 14 shows an example of a system using a total of four types of dots that can form two types of dots, Bk ink, large dots and small dots. As described above, the tone correction processing output values are checked in order from the dot with the highest optical density, and the combination to be quantized is determined.

  When this embodiment is generalized, quantized data used to perform recording by forming three or more types of dots having the same hue such as monochrome and cyan and having different optical densities is generated. In this case, when quantizing a plurality of multi-value data to obtain quantized data having a smaller number of gradation values, the quantization obtained by associating a plurality of multi-value data with each other by a plurality of multi-value data Quantized data is obtained by referring to a table storing data. Then, according to the value of multi-value data corresponding to three or more types of dots, a table to be referred to by multi-value data corresponding to two types of dots is selected from among the multi-value data corresponding to the three or more types of dots. .

(Embodiment 3)
The present invention can also be applied to a mode in which ink other than the achromatic color ink shown in the above-described example is used, that is, a mode in which recording is performed using a plurality of dots having substantially the same hue but different optical densities even for chromatic colors. The present invention can be applied.

  For example, in addition to C (large dot), M (large dot), Y, Bk, c (small dot), and m (small dot), cyan medium dot (C ′) and magenta medium dot (M ′) The present invention can be applied to a form in which recording is performed. That is, for cyan and magenta, recording is performed using three types of dots, large dots, medium dots, and small dots.

In the configuration shown in FIG. 7, the RGB data is separated into eight types of colors CMYKcmC′M ′ by the color separation processing unit 707, and gradation correction is performed on the data of each color by the gradation correction processing unit 708. Here, the input to the color separation processing unit 707 has 256 gradations for each of R, G, and B, and there are 256 ³ types of inputs. Of these, for example, cyan component = 255-R Focus only on ingredients.

  FIG. 15 is a diagram showing the relationship between the input level of the cyan component and the output value from the gradation correction processing unit 708, and is the same diagram as FIG. As shown in FIG. 15, cyan large dots, medium dots, and small dots are used in accordance with the input level of the cyan component. Whether the quantization processing unit 709 determines whether the large cyan dot is 0 or not, performs the quantization with the combination of the large dot and the medium dot, or performs the quantization with the combination of the medium dot and the small dot. Switch. The same applies to magenta.

(Other embodiments)
Each of the above embodiments has been described as an example in which the quantization processing according to the present invention is executed by the printer driver of the host device, but this quantization processing may be performed in the printing apparatus. In any case, the apparatus that executes the quantization processing according to the present invention constitutes an image processing apparatus.

(Still another embodiment)
The present invention can also be realized by a program code that realizes the functions of the above-described embodiments and that realizes the procedures of the flowcharts shown in FIGS. 9, 13, and 14, or a storage medium storing the program code. The program code stored in the storage medium is also read and executed by a computer (or CPU or MPU) of the system or apparatus. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the storage medium storing the program code constitutes the present invention.

  As a storage medium for supplying the program code, for example, a floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, magnetic tape, nonvolatile memory card, ROM, or the like is used. be able to.

  In addition, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also the OS running on the computer performs an actual process based on the instruction of the program code. Part or all may be performed.

  Further, after the program code is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the CPU or the like may perform part of the actual processing or based on the instruction of the program code. You may do everything.

702 Determination unit 703 Gray scale conversion unit 704, 707 Color separation processing unit 705, 708 Tone correction processing unit 706, 709 Quantization unit 710 Dot arrangement patterning processing unit

Claims (14)

An image processing apparatus for recording a first dot optical density are different, the image is formed on the second dot and the third dot over the recording medium in the same color,
Based on the color component data of the image, first multi-value data corresponding to the first dot, second multi-value data corresponding to the second dot, and third multi-value data corresponding to the third dot Color separation means for generating
If none of the values indicated by the first, second, and third multi-value data is 0, the first multi-value data and the second multi-value data are used and the third multi-value data is not used. Generating the first quantized data for forming one dot and forming the second dot using the first multi-value data and the second multi-value data without using the third multi-value data; the image processing apparatus characterized by obtaining ingredients and a quantization means for generating second quantized data. The quantization means quantizes the input first multi-value data and second multi-value data using a table in which first quantized data and second quantized data to be output are determined. The image processing apparatus according to claim 1. The image processing apparatus according to claim 1, wherein the quantization unit performs quantization by executing error diffusion processing. The quantization means generates third quantized data for forming the third dots using the third multi-value data without using the first multi-value data and the second multi-value data. The image processing apparatus according to claim 1, wherein: 5. The image processing apparatus according to claim 1 , wherein the second dot has a lower optical density than the first dot and a higher optical density than the third dot . 6. The image processing apparatus according to claim 1, wherein the second dots and the third dots have different sizes. When the value indicated by the first multi-value data is 0, the quantization means uses the second multi-value data and the third multi-value data to change the second dot without using the first multi-value data. A fourth quantized data for forming is formed, and a fifth dot for forming the third dot by using the second multi-value data and the third multi-value data without using the first multi-value data. The image processing apparatus according to claim 5, wherein quantized data is generated. First dot different optical density respectively in the same color, an image processing method for recording an image by forming a second dot and the third dot on the recording medium,
Based on the color component data of the image, first multi-value data corresponding to the first dot, second multi-value data corresponding to the second dot, and third multi-value data corresponding to the third dot A color separation process to generate,
If none of the values indicated by the first, second, and third multi-value data is 0, the first multi-value data and the second multi-value data are used and the third multi-value data is not used. Generating the first quantized data for forming one dot and forming the second dot using the first multi-value data and the second multi-value data without using the third multi-value data; an image processing method characterized by chromatic quantization step, a generating a second quantized data. In the quantization step, the input first multi-value data and the second multi-value data are quantized using a table in which the first quantized data and the second quantized data to be output are determined. The image processing method according to claim 8. The image processing method according to claim 8 or 9, wherein in the quantization step, quantization is performed by executing error diffusion processing. In the quantization step, generating the third quantized data for forming the third dots using the third multi-value data without using the first multi-value data and the second multi-value data. The image processing method according to claim 8, wherein: The image processing method according to claim 8, wherein the second dot has an optical density lower than that of the first dot and higher than that of the third dot. The image processing method according to claim 8, wherein the second dots and the third dots have different sizes. In the quantization step, when the value indicated by the first multi-value data is 0, the second multi-value data and the third multi-value data are used and the second dot is not used without using the first multi-value data. A fourth quantized data for forming is formed, and a fifth dot for forming the third dot by using the second multi-value data and the third multi-value data without using the first multi-value data. 14. The image processing method according to claim 12, wherein quantized data is generated.

Images